Computer implemented methods for the automated analysis or use of data, including use of a large language model

Abstract

Methods are provided, such as a method of interacting with a large language model (LLM), including the step of a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, to provide new context data for the LLM, in order to improve the output, such as continuation text output, generated by the LLM in response to a prompt; and such as a method of interacting with a LLM, including the step of providing continuation data generated by the LLM to a processing system that uses a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language, in which the processing system is configured to analyse the continuation output generated by the LLM in response to a prompt to enable an improved version of that continuation output to be provided to a user. Related computer systems are provided.

The invention claimed is: 
     
       1. A computer-implemented method of interacting with a large language model (LLM), including the steps of:
 (a) the LLM processing first input data to the LLM to generate first output from the LLM based on the first input data to the LLM; 
 (b) a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, in which semantic nodes are represented in the machine-readable language, the semantic nodes including semantic links between semantic nodes wherein the semantic links are themselves semantic nodes, in which each semantic node denotes one specific meaning, in which a combination of semantic nodes defines a semantic node, in which expressions in the machine-readable language are nestable, in which the first output from the LLM is represented in the machine-readable language, in which reasoning steps are represented in the machine-readable language to represent semantics of the reasoning steps, in which computation units are represented in the machine-readable language; 
 (c) the processing system verifying the first output from the LLM using the reasoning steps, the computation units and the semantic nodes, and 
 (d) providing second input data to the LLM, including the verified first output from the LLM in order to generate improved first output from the LLM, wherein the improved first output from the LLM is generated by the LLM in response to the second input data. 
 
     
     
       2. The method of  claim 1 , in which the second input data is a correction and/or a change of continuation text output generated by the LLM. 
     
     
       3. The method of  claim 2 , in which the LLM provides output based on the inputted correction and/or change of continuation text output generated by the LLM. 
     
     
       4. The method of  claim 1 , in which the second input data is provided as at least part of a prompt to the LLM. 
     
     
       5. The method of  claim 1 , in which factual accuracy and/or factual scope of the first output is improved by the second input data. 
     
     
       6. The method of  claim 1 , in which internal, logical self-consistency (e.g. correct time ordering of events) of the first output is improved by the second input data. 
     
     
       7. The method of  claim 1 , in which correspondence of the first output generated by the LLM to how people understand the real world or reason in the real world is improved by the second input data. 
     
     
       8. The method of  claim 1 , in which bias in the first output is reduced by the second input data. 
     
     
       9. The method of  claim 1 , in which the second input data provided to the LLM includes dynamic or real-time information. 
     
     
       10. The method of  claim 1 , in which the second input data provided to the LLM includes reasoned text, e.g. text derived from a non-statistical reasoning process. 
     
     
       11. The method of  claim 1 , in which the LLM is answering a question and the second input data provided to the LLM is an answer to that question. 
     
     
       12. The method of  claim 1 , in which second input data text provided to the LLM is labelled with a level of certainty or uncertainty, or trust or lack of trust. 
     
     
       13. The method of  claim 1 , in which second input data text provided to the LLM is labelled with a level of brevity. 
     
     
       14. The method of  claim 1 , in which second input data text provided to the LLM is labelled with a level of formality. 
     
     
       15. The method of any  claim 1 , in which second input data text provided to the LLM is labelled with a use or non-use of profanity. 
     
     
       16. The method of  claim 1 , in which second input data text provided to the LLM is labelled with an age or other details of a person being addressed by the LLM. 
     
     
       17. The method of  claim 1 , in which the first output from the LLM is at least partially translated to the machine-readable language and is analysed for factual inaccuracies or other contradictions. 
     
     
       18. The method of  claim 1 , in which a classifier operates to identify when a prompt is likely to result in a continuation output where accuracy is important, and/or when accuracy is important in the continuation output, and to then use the processing system to improve factual accuracy and/or factual scope of that continuation output. 
     
     
       19. The method of  claim 1 , in which the first output from the LLM is a partial continuation, namely an output made before the LLM has stopped generating or whilst the LLM is still generating. 
     
     
       20. The method of  claim 1 , when used to improve one or more parameters of the first output of the LLM: factual accuracy and/or factual scope of the first output; internal, logical self-consistency (e.g. correct time ordering of events) of the first output; bias reduction or removal in the first output; inclusion of dynamic or real-time information. 
     
     
       21. The method of  claim 1 , when used to improve one or more parameters of the first output of the LLM: level of formality, level of brevity (e.g. briefer when the language is to be spoken), suitability for speaking via a text to speech system, other style language, level of certainty. 
     
     
       22. The method of  claim 1 , in which the LLM is a generative AI based system. 
     
     
       23. The method of  claim 1 , in which the LLM is an autoregressive language model, such as a Generative Pre-trained Transformer. 
     
     
       24. The method of  claim 1 , when used for any of the following: generation of program code, solution of any problem that can be described in natural language, generation of poetry, lyrics, creative writing, generation of other forms of writing such as essays, summaries of knowledge, summaries of longer texts, essays, scientific papers; question answering; internet search. 
     
     
       25. The method of  claim 1 , including identifying output content that breaches predefined policies by translating the first output into the machine-readable language before it is displayed to the user and then checking the machine-readable language representation against the predefined policies. 
     
     
       26. The method of  claim 1 , including the step of training the LLM on output from a processing system using a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language. 
     
     
       27. The method of  claim 1 , including a computer implemented method of augmenting search results, including the steps of:
 (i) receiving a search query; 
 (ii) an (e.g. internet) search engine processing the search query to generate (e.g. internet) search results; 
 (iii) transforming the search results into a structured, machine-readable representation of data that conforms to a machine-readable language, such as a universal language; 
 (iv) processing the transformed search results using a processing system which uses the structured, machine-readable representation of data that conforms to the machine-readable language, such as a universal language, to produce output; 
 (v) supplying the output of step (iv) as input to the large language model (LLM), and the LLM generating continuation data output in response to the input. 
 
     
     
       28. The method of  claim 27 , including the step of outputting to a user interface the LLM generated continuation data output. 
     
     
       29. The method of  claim 1 , wherein the improved first output from the LLM is continuation text output. 
     
     
       30. The method of  claim 1 , wherein the machine-readable language is a universal language.